Qwen3.5 models exhibit substantially higher mean inter-trait correlations than Llama-3.1-8B and Gemma-3-27B (mean off-diagonal 0.215 and 0.146 vs 0.090 and 0.078), indicating greater trait entanglement.

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• The difficulty boundary for truth directions replicates across all four tested models (Llama-3.2-3B, Llama-3.1-8B, Gemma-2-2b, Gemma-2-9b); generalization to F3-F5 remains consistently low regardless of model size or family.finding0.816
  Establishes generalizability of the core difficulty-boundary finding across model families.
• Qwen 2.5 7B has higher Elo scores for 'methodical' and 'formal' traits compared to Llama 3.1 8B which prefers 'colloquial'finding0.815
  Model-specific baseline personality difference revealed by revealed preferences experiment
• Pairwise correlation of role loadings on PC2 is 0.89 between Qwen and Llama; Gemma differs (similarity <0.61) from others on PC2finding0.805
  Characterizes model similarities and differences in secondary persona dimensions
• Qwen3.5-9B and Qwen3.5-27B show very high SP persona rank correlation (rho=0.959, p=4.32e-14) but diverge sharply on AS vulnerability.finding0.802
  Scale effects within Qwen3.5 family on different imbuing methods
• Gemma-3-27B and Qwen3.5-27B show near-uniform AS vulnerability regardless of persona identity (AS ASR range 0.095-0.117 on Gemma-3-27B, 0.015-0.052 on Qwen3.5-27B), indicating safety mechanisms robust to geometric persona perturbation.finding0.794
  Qualitatively different defense profile compared to Llama-3.1-8B
• The generalization improvement from explicit instructions observed in Llama models (A1-A3 to F0-F2) is more pronounced for F3-F5 to F0-F2 in Gemma models.claim0.789
  Shows the instruction effect, while shifting geometry, may not produce consistent generalization effects across model families.
• Pairwise structural correlations of inter-trait cosine similarity matrices across all four architectures range from r=0.898 to r=0.986 (all p<0.001), indicating cross-architecture geometric preservation.finding0.784
  Cross-architecture geometric invariance of Big Five steering vectors
• Trait space requires 4 dimensions (Gemma, Qwen) and 7 dimensions (Llama) to explain 70% of variance, with distinctive PC1 spanning conscientious to impulsive traitsfinding0.782
  Corroborates role space findings using traits; shows PC1 also captures Assistant-ness in trait space